Yeah, the question is, why the fuck do we need the olpc project then? The hardware is already out there. Do you want a tougher *Pad? add a rubber case.
Any device for children has an exacting list of requirements for hardware, software, localization, educational curricula, infrastructure for administration / distribution / repair and all the bureaucracy that goes along with it.
I'm sure you could stick a rubber case on an aPad but that doesn't make it fit for the task.
Because the OS in this case is a stripped down Red Hat distribution with X11 on top and it's not suitable for a tablet. It uses too much memory, takes too long to start and X11 is utterly unsuitable for touch. I expect Sugar is also woefully unsuited for touch devices and would require major work, especially if it has to be ported.
There are embedded Linux dists which would be more suitable, or even Android which is built from the ground up for smart phones and touch devices.
There are already awesome tablets like the aPad that exist right now and retail for less than 200 dollars. I'm sure you could drive them below 100 if you built enough and bought them altogether.
The aPad isn't awesome, but it does exist and is a functional tablet device that can be had for $100. I think it demonstrates that OLPC could produce a rugged 7" tablet for less than $100 and it would be great for kids. My own feeling however is that the hardware is the easy bit and it will be the software which will prove difficult. I don't think much can be salvaged from OLPC 1.x, not the OS nor the sofware without a substantial rewrite to support touch based input.
Don't be silly Mr AC. I call Windows Windows regardless of what other software I run on it. I call OS X OS X regardless of what other software I run on it. And I call Linux Linux regardless of what other software I run on it. A single word is a perfectly adequate way to refer to the OS.
The arguments made by the FSF that it should be called GNU/Linux are incredibly weak and incredibly selfish. There is no idealism at work in laughing at those arguments. Linux is the OS and Linux is a name everyone understands.
And in case you think the same should not apply to the Linux kernel in some situations - wrong. Android is Android even though it has a Linux kernel in it. WebOS is WebOS even though it has a Linux kernel in it. People who advance the theory that Android should be called BSD/Linux or Linux/BSD/Android or even Linux/BSD/GNU/Android should be laughed at with equal force to those insisting on GNU/Linux.
Are there any laws governing what you can legally name your organization?
In Britain there are naming rules that require names be unique, none infringing, don't imply a connection to government or royalty, are not offensive, or confusing (e.g. Limited ltd). There are certain additional rules when you include words like Vet, Doctor, Solicitor etc. in your company name.
Even with the rules it doesn't stop some scummy ambulance chasing companies trying to pass themselves off as official sounding accident boards and such like.
One of the reasons why big business loves Windows and isn't that interested in Linux other than maybe Red Hat is because if things go horribly wrong, there's somebody with deep pockets to sue. What Microsoft is offering here is a classic part of their business plan... if somebody comes up with a submarine patent they'll take the legal pain so their customers don't have to.
Are you saying Google, HP, Nokia or Apple who all have their own phone operating systems don't have deep pockets? In two cases the Linux kernel is powering the OS and in another it's a BSD variant.
I think this would be a reasonable solution. By default all preinstalled apps are trusted. If you install an app from marketplace, it's untrusted which either blocks outright or prompts when it tries to do something iffy such as dial a number, obtain your IMEI or whatever. In the case that you do trust an app, you can elevate its trust level and the phone will stop nagging you.
GNU is just part of a regular Linux distribution, the sum of which people quite happily refer to by the simple moniker Linux, or if more context is required Red Hat Linux etc.
It strikes me as immature and selfish that anyone insist any dist be called GNU/Linux. There are substantial non-FSF parts to all mainstream dists. Parts from Apache, X11, Mozilla, Sun, Novell etc. Parts that make the dist useful for something and all of which deserve credit. But unless we intend to say Mozilla/Apache/Perl/Aladdin/X11/MIT/BSD/RedHat/Novell/Sun/Trolltech/GNU/etc./Linux, just plain old Linux is perfectly simple understand and is also the right thing.
Insisting on calling it GNU/Linux is disingenuous, dishonest, nonsensical and probably motivated by sour grapes. If the FSF doesn't like people running Linux and using the many non-FSF contributions that make it useful, perhaps they should get cracking on Hurd.
Absolutely. I think this is the biggest failing of the Android security model. In principle its fine, but the permissions it requests are too broad or ambiguous and once you grant them up front you can never change them afterwards. I believe that regardless of what permissions the app says it needs, users (possibly by default) should be protected by something similar to UAC where they either grant or deny access to their private data. The dialog could also allow them to bestow complete trust in the app if they never want to see nag screens again, but by default more security is needed.
I expect it will not be long before smart phones get 3D displays. Sharp has been touting them recently at smaller sizes which suggests they're after that market. I have no issue with the concept of a 3D display as long as it toggles back to a decent 2D display and doesn't affect the screen in high sunlight and so on. A phone which expects users to exactly hold it in the horizontal at a fixed distance to see the effect is going to suck unless it has a decent display for the 99% of other times when 3D is not required.
3D is still a gimmick. I expect sooner or later someone will produce a decent 3D display that works from a wide viewing angle from any orientation. Until that time, devices that hype 3D should really be treated with a grain of salt.
Capture from multiple sources and check for differences. It should be easy enough to find watermarks and not terribly difficult to edit them out automatically. If there's a unique identifier, it can't be in two sources.
And who is going to have two raw streams to compare? If the sources are compressed then any differences are meaningless since there will be so many. Even if two people had exactly the same codec and exactly the same settings the files not be the same unless they started at the exact same frame.
Besides which, making imperceptible watermarks than can survive arbitrary lossy compression techniques is hard. If a person can't see a certain detail, a good video compression algorithm should throw it out. Just saying.
Watermarks are designed to be robust which is the point. You'd probably have to degrade the image so much that it wouldn't be worth bothering at all and even then you wouldn't know for sure unless you have a tool to detect the watermarks.
A much more interesting class break would be if someone cracked DVB-CSA. Then people could rip content straight from the cable / satellite / receiver without any trace of where it came from.
The reason this is useful is not for bluray, it is for first-run broadcast content.
I expect most modern STBs watermark the content leaving the TV in addition to HDCP protection. Therefore anyone stupid enough to release content they've captured on P2P can probably kiss their service goodbye as well as opening up the possibility of prosecution. Just saying.
2D to 3D conversion is rapidly becoming the modern day equivalent of colorization of black and white films. I wonder how long before "now in 3D!" becomes as distasteful as "now in color!".
The conversion process is basically the computational equivalent of vacuum forming the 2D image over a depth map. Depending on the fidelity of the models that generate the depth map, the effect will vary between passable and looking like cardboard cutouts in a diorama.
From a technical point of view it will be interesting to see how Lucasfilm deal with the conversion. I think the conversion will be abysmal in the first 3 movies with lots of crappy dust & particle effects tossed in and some reshot CG. Potentially the 3D could be more passable on the prequels depending on how far they go. Most of the prequel scenes are pure CG or composite live action and CG so in theory all these scenes could be rerendered. I think in practice though that probably only a fraction will get the full treatment and the remainder will under go a 2D + depth map process.
Isn't that an opportunity for arbitrage? Gold doesn't change price every ten minutes exactly, so if there's a spike in the right direction, buy before the ATM updates its prices...
Except that the dispenser most likely calculates the price as market price + obscene markup. You'd have to cause a huge spike in the market and then employ a gang to simultaneously buy up gold from all the ATMs at the cheaper price and somehow prolong the spike long enough to sell your bullion at the raised price in order to benefit.
I think it would be easier to hire yourself a commodity trader to do all of the above and forget about the ATM entirely.
As is usually the case, the technology itself isn't really good or bad, but what people do with it can be. And people, as a rule, are decidedly good at making technology do bad things.
Exactly. People complain that Flash performance sucks, probably while having 10 tabs open each with 2 or 3 flash plugin. The plugin is a victim of its own success. When an ad blocker is employed to cut down the amount of crap (surprise) Flash performance is a lot better.
Another weird thing is when people complain of Flash they usually follow up with some comment of how HTML 5 is going to save us all. There is an implied assumption that HTML 5 + JS somehow offers better performance than Flash in the same role. Boy are people going to be disappointed.
It won't be better since it will be faced with the same technical issues faced by Flash, such as software vs hardware acceleration for decoding, converting video to RGB, timing critical animation, tweening etc. When the tools invariably appear that spit out the HTML 5 equivalent of today's Flash content, performance is going to go down the toilet. At least with Flash, the plugin has the option to run in a separate thread, or only disturb the browser when it's done rendering. With HTML 5 virtually everything in the page runs in a single thread (with the exception of web workers) so a couple of inserted JS animations could easily bring a browser to its knees.
"Lots of us think your utter submission to your governments, preference for the safety of lawbreakers over personal self-defense, and general sheeple tendencies aren't admirable either. You've traded freedom for (the perception of) security as is your right, but that only works in certain situations and assumes benign government."
Anyone who says "sheeple" in earnest should require a psychiatrists evaluation before being allowed to own a gun.
... it doesn't work very well, there aren't any games out for it yet, they're late to market and it looks like a sex toy for a Dalek.
Actually the hardware works extremely well and is getting high praise. The launch lineup runs the gamut from excellent to tacked on. As with any launch lineup, you pick and choose the best titles (Tumble, Sports Champions) or wait for the next wave.
By all accounts, the Move should blow the basic Wii controller out of the water and be at least par with the Motion Plus.
And it does. It's an extremely precise and accurate controller and the lag is not really noticeable. It is so accurate that the Start the Party game is able to overlay a sword / bat / etc exactly over the top of your camera image.
Whether this means anything in the long term is another matter.
I think Google should just kill Chrome OS. Port the web application framework into Android and junk the rest. Chrome OS is never going to fly and it's just sending mixed messages to developers and consumers about Google's intentions. This dithering is already affecting Android - many forthcoming tablet devices won't be able to meet the CTS / CDD compliance tests and therefore won't be able to ship the standard marketplace app. I certainly wouldn't want to buy a tablet unless it did. Maybe Android 3.0 will resolve these issues, but I certainly wouldn't want to buy a 2.2 device unless I knew for certain an update was forthcoming later.
There is not a chance in hell that I'd buy a blu-ray unless I could store and back-up the contents on a regular media server. I hate all those little plastic boxes, and I also hate the anti-piracy messages and studio branding.
Well don't forget you wouldn't actually be seeing the ads. Your requests would go through Haystack which would decrypt the payload inside the ad's cookies / shared object and show that to you.
Of course it's all speculation how Haystack works since they've hardly been forthcoming. Perhaps it uses another kind of backchannel.
Slashdot Mirror
Any device for children has an exacting list of requirements for hardware, software, localization, educational curricula, infrastructure for administration / distribution / repair and all the bureaucracy that goes along with it.
I'm sure you could stick a rubber case on an aPad but that doesn't make it fit for the task.
There are embedded Linux dists which would be more suitable, or even Android which is built from the ground up for smart phones and touch devices.
The aPad isn't awesome, but it does exist and is a functional tablet device that can be had for $100. I think it demonstrates that OLPC could produce a rugged 7" tablet for less than $100 and it would be great for kids. My own feeling however is that the hardware is the easy bit and it will be the software which will prove difficult. I don't think much can be salvaged from OLPC 1.x, not the OS nor the sofware without a substantial rewrite to support touch based input.
A quick google shows how absurd virtually everything you wrote is.
The arguments made by the FSF that it should be called GNU/Linux are incredibly weak and incredibly selfish. There is no idealism at work in laughing at those arguments. Linux is the OS and Linux is a name everyone understands.
And in case you think the same should not apply to the Linux kernel in some situations - wrong. Android is Android even though it has a Linux kernel in it. WebOS is WebOS even though it has a Linux kernel in it. People who advance the theory that Android should be called BSD/Linux or Linux/BSD/Android or even Linux/BSD/GNU/Android should be laughed at with equal force to those insisting on GNU/Linux.
In Britain there are naming rules that require names be unique, none infringing, don't imply a connection to government or royalty, are not offensive, or confusing (e.g. Limited ltd). There are certain additional rules when you include words like Vet, Doctor, Solicitor etc. in your company name.
Even with the rules it doesn't stop some scummy ambulance chasing companies trying to pass themselves off as official sounding accident boards and such like.
Are you saying Google, HP, Nokia or Apple who all have their own phone operating systems don't have deep pockets? In two cases the Linux kernel is powering the OS and in another it's a BSD variant.
I think this would be a reasonable solution. By default all preinstalled apps are trusted. If you install an app from marketplace, it's untrusted which either blocks outright or prompts when it tries to do something iffy such as dial a number, obtain your IMEI or whatever. In the case that you do trust an app, you can elevate its trust level and the phone will stop nagging you.
It strikes me as immature and selfish that anyone insist any dist be called GNU/Linux. There are substantial non-FSF parts to all mainstream dists. Parts from Apache, X11, Mozilla, Sun, Novell etc. Parts that make the dist useful for something and all of which deserve credit. But unless we intend to say Mozilla/Apache/Perl/Aladdin/X11/MIT/BSD/RedHat/Novell/Sun/Trolltech/GNU/etc./Linux, just plain old Linux is perfectly simple understand and is also the right thing.
Insisting on calling it GNU/Linux is disingenuous, dishonest, nonsensical and probably motivated by sour grapes. If the FSF doesn't like people running Linux and using the many non-FSF contributions that make it useful, perhaps they should get cracking on Hurd.
Absolutely. I think this is the biggest failing of the Android security model. In principle its fine, but the permissions it requests are too broad or ambiguous and once you grant them up front you can never change them afterwards. I believe that regardless of what permissions the app says it needs, users (possibly by default) should be protected by something similar to UAC where they either grant or deny access to their private data. The dialog could also allow them to bestow complete trust in the app if they never want to see nag screens again, but by default more security is needed.
3D is still a gimmick. I expect sooner or later someone will produce a decent 3D display that works from a wide viewing angle from any orientation. Until that time, devices that hype 3D should really be treated with a grain of salt.
And I thought the PSP was expensive. How do they keep their prices so low and still make a profit?
And who is going to have two raw streams to compare? If the sources are compressed then any differences are meaningless since there will be so many. Even if two people had exactly the same codec and exactly the same settings the files not be the same unless they started at the exact same frame.
Besides which, making imperceptible watermarks than can survive arbitrary lossy compression techniques is hard. If a person can't see a certain detail, a good video compression algorithm should throw it out. Just saying.
Watermarks are designed to be robust which is the point. You'd probably have to degrade the image so much that it wouldn't be worth bothering at all and even then you wouldn't know for sure unless you have a tool to detect the watermarks.
A much more interesting class break would be if someone cracked DVB-CSA. Then people could rip content straight from the cable / satellite / receiver without any trace of where it came from.
I expect most modern STBs watermark the content leaving the TV in addition to HDCP protection. Therefore anyone stupid enough to release content they've captured on P2P can probably kiss their service goodbye as well as opening up the possibility of prosecution. Just saying.
The conversion process is basically the computational equivalent of vacuum forming the 2D image over a depth map. Depending on the fidelity of the models that generate the depth map, the effect will vary between passable and looking like cardboard cutouts in a diorama.
From a technical point of view it will be interesting to see how Lucasfilm deal with the conversion. I think the conversion will be abysmal in the first 3 movies with lots of crappy dust & particle effects tossed in and some reshot CG. Potentially the 3D could be more passable on the prequels depending on how far they go. Most of the prequel scenes are pure CG or composite live action and CG so in theory all these scenes could be rerendered. I think in practice though that probably only a fraction will get the full treatment and the remainder will under go a 2D + depth map process.
Except that the dispenser most likely calculates the price as market price + obscene markup. You'd have to cause a huge spike in the market and then employ a gang to simultaneously buy up gold from all the ATMs at the cheaper price and somehow prolong the spike long enough to sell your bullion at the raised price in order to benefit.
I think it would be easier to hire yourself a commodity trader to do all of the above and forget about the ATM entirely.
You have split the child in two. You have gained 13 exp. Welcome to level 2!
Exactly. People complain that Flash performance sucks, probably while having 10 tabs open each with 2 or 3 flash plugin. The plugin is a victim of its own success. When an ad blocker is employed to cut down the amount of crap (surprise) Flash performance is a lot better.
Another weird thing is when people complain of Flash they usually follow up with some comment of how HTML 5 is going to save us all. There is an implied assumption that HTML 5 + JS somehow offers better performance than Flash in the same role. Boy are people going to be disappointed.
It won't be better since it will be faced with the same technical issues faced by Flash, such as software vs hardware acceleration for decoding, converting video to RGB, timing critical animation, tweening etc. When the tools invariably appear that spit out the HTML 5 equivalent of today's Flash content, performance is going to go down the toilet. At least with Flash, the plugin has the option to run in a separate thread, or only disturb the browser when it's done rendering. With HTML 5 virtually everything in the page runs in a single thread (with the exception of web workers) so a couple of inserted JS animations could easily bring a browser to its knees.
I guess that means he can have 28 chicks at once!
Anyone who says "sheeple" in earnest should require a psychiatrists evaluation before being allowed to own a gun.
Actually the hardware works extremely well and is getting high praise. The launch lineup runs the gamut from excellent to tacked on. As with any launch lineup, you pick and choose the best titles (Tumble, Sports Champions) or wait for the next wave.
And it does. It's an extremely precise and accurate controller and the lag is not really noticeable. It is so accurate that the Start the Party game is able to overlay a sword / bat / etc exactly over the top of your camera image.
Whether this means anything in the long term is another matter.
I think Google should just kill Chrome OS. Port the web application framework into Android and junk the rest. Chrome OS is never going to fly and it's just sending mixed messages to developers and consumers about Google's intentions. This dithering is already affecting Android - many forthcoming tablet devices won't be able to meet the CTS / CDD compliance tests and therefore won't be able to ship the standard marketplace app. I certainly wouldn't want to buy a tablet unless it did. Maybe Android 3.0 will resolve these issues, but I certainly wouldn't want to buy a 2.2 device unless I knew for certain an update was forthcoming later.
Well buy AnyDVD HD and you can.
Of course it's all speculation how Haystack works since they've hardly been forthcoming. Perhaps it uses another kind of backchannel.